Frequently Asked Questions about the UBOS MCP Server

Q: What is the UBOS MCP Server?

A: The UBOS MCP (Model Context Protocol) Server is a tool that allows developers to access and query the AI Incident Database (AIID) directly within VS Code. It acts as a bridge, enabling AI models to interact with external data sources.

Q: What is the AI Incident Database (AIID)?

A: The AIID is a comprehensive database of AI incidents, including failures, biases, and adversarial attacks. It provides valuable insights for understanding the risks associated with AI systems.

Q: How does the UBOS MCP Server help with responsible AI development?

A: By providing easy access to incident data, the MCP Server empowers developers to proactively identify potential risks, mitigate biases, and improve the safety and reliability of their AI systems.

Q: What are the key features of the UBOS MCP Server?

A: Key features include seamless VS Code integration, AIID access, flexible querying, customizable data fields, filtering, sorting, and data transformation capabilities.

Q: How do I install the UBOS MCP Server?

A: You can install the MCP Server using npm: npm install -g aiid-mcp. Then, configure VS Code by adding the MCP server configuration to your settings.json file and restart VS Code.

Q: What output formats are supported by the UBOS MCP Server?

A: The MCP Server supports both JSON and CSV output formats.

Q: Can I filter and sort the data retrieved from the AIID?

A: Yes, you can filter and sort incident data based on various criteria, such as date, severity, and affected system.

Q: How does pagination support work?

A: Pagination support allows you to control the result size and offset for large datasets, enabling efficient handling of large amounts of data.

Q: Where can I find more information about the Model Context Protocol (MCP)?

A: You can find more information about the MCP on the official Model Context Protocol Documentation: https://github.com/microsoft/modelcontextprotocol.

Q: How can I contribute to the UBOS MCP Server project?

A: Contributions are welcome! Please feel free to submit a Pull Request on the project’s GitHub repository.